Be Careful What You Incentivize: Outcomes vs. Behaviors

There’s a pretty interesting story I came across in a strategy class offered by one of the professors at Harvard Business a few years ago. The topic for this particular session was corporate scorecards. We were examining how top-level metrics get translated into lower-level behaviors.

It has been fairly en vogue for the past several years to implement these balanced scorecards. The general theory is that if you understand the small set of leading and lagging indicators that drive your business, you can better tune your decisions and execute more effectively to meaningfully move the needle. And if employees understand how their specific actions contribute to these metrics, they are in a better position to make the best on-the-ground decisions for the company. And so companies have been fairly diligent in outlining what their top-line metrics are, and then connecting these through management objectives (MBOs) through the ranks of their employees.

But you need to be careful how you create these metrics.

The Harvard B-school professor told a story about a fast food chain. They had identified customer service as an important metric. One of the proxies for customer service was the time it took their staff to serve customers in the drive-thru. So they implemented a Waiting Time metric at all of their stores.

When a customer pulled up in the drive-thru, a timer that was visible to both the customer and the server would flash 0:00 and then begin timing the visit. The store had determined that all visits to the counter should take no more than 90 seconds. When a visit lasted over 90 seconds, it would record. Stores would then report their daily, weekly, and monthly statistics back to corporate.

On one particular visit, a customer placed an order and drove around to the window. When he arrived at the window, he told the server that he had forgotten to order a side. She said she would gladly take care of it; it would be another order. So she handed him his previous order (without side), and then she asked him to drive forward 15 feet and then back his car up to the window again. When he asked why he had to do that, she explained that the timer would reset to 0:00 when he pulled forward, and so when he pulled back he could get his side without having triggered the length-of-visit response.

From the bleachers, this one seems dead stupid. It is easy to see that the woman at the drive-thru window made a poor decision in pursuit of a bad metric. But it raises an interesting point: when you set metrics for your organization, are you certain how those metrics and incentives are going to play out?

In a different well-known example, a software company decided it would provide incentives to its test teams based on how many bugs they found. It didn’t take long for the clever engineers to figure out that if they split those incentives with the software developers, they could maximize returns. The software engineers would intentionally introduce defects, and they would split the proceeds with their QA counterparts.

Again, the issue is that the metric and incentives were aligned less to the outcome and more to specific behaviors. The second example is certainly more sketchy in that it involves people intentionally gaming the system, but the point remains: if you do not carefully consider the outcomes you are trying to drive towards, your incentives might end up feeding into behaviors that are not ideal for your company.

In the first example, the outcome that the company wants is better customer satisfaction. When that overarching objective is lost in the translation to metrics, the individual becomes more attached to the metric than the end goal. In the second example, the outcome is better software quality (or better end user experience). Again, when that is lost in translation, the metric dominates.

Successful leaders of organizations of any size need to make sure they do not rely on metrics at the expense of communicating intent. It is very easy to think that simply having a measurable objective is enough to align the organization. But failing to communicate the intent behind the metrics leaves an unengaged workforce to interpret the meaning on their own. And when this happens, the outcome is anything but certain.

Instead, effective leaders will communicate the intent as part of their daily dialogue with teams. Their generals will communicate the intent as part of their daily dialogue. The corporate vernacular needs to be filled with everyday occurrences of the actual objective. The end goal needs to be a daily conversation had at all levels of the organizations. This conversational backdrop provides the context for the metrics being discussed. And the incentive programs need to feed off the metrics being collected but align to the overarching goal. If there is a disconnect anywhere along the way, leaders beware — the results can be quite unpredictable.

So if you are a leader in a metrics-driven organization, check yourself:

• Can you articulate in plain english what outcomes are important to your company?
• Can you describe how your actions and the actions of your team support those objectives?
• Do you understand implicitly how the metrics reflect those actions?
• And, most importantly, can your team do the same?

See more at: http://www.plexxi.com/2013/10/be-careful-what-you-incentivize-outcomes-vs-behaviors/#sthash.zHVYl0gX.dpuf